臨床大數據分析與挖掘——基於Python和機器學習的臨床決策

孫麗萍,張良均

買這商品的人也買了...

商品描述

本書不僅講解了機器學習基本原理和基本方法,而且通過大量醫療領域的案例實現對醫療健康數據的處理和分析,能夠在很大程度上輔助醫護人員進行臨床決策。通過本書學習,讀者不僅能夠掌握機器學習算法建模前的數據準備、篩選構造機器學習算法指標的特徵工程、不同類別的機器學習算法,還能夠掌握臨床診療數據、電子病歷檔案數據及影像數據等多源異構數據的處理方法,以及醫療圖像、文本等數據的讀取、預處理、可視化等知識。同時,本書還介紹了具有開源、去編程化的TipDM 數據挖掘建模平臺,通過拖曳的圖形化操作就能實現數據分析的全流程。本書可以作為醫學類院校數據科學與大數據技術專業的核心課程教材,以及醫工專業的專業核心課程或選修課程教材。在此基礎上,還可以作為臨床、口腔、醫技、檢驗、影像、公共衛生等醫學類專業進階層次的專業限選課程或拓展課程的教材。

目錄大綱

第1 章機器學習 ··············································································································1
1.1 機器學習簡介·······································································································1
1.1.1 機器學習的概念······························································································1
1.1.2 機器學習的應用領域························································································1
1.2 機器學習通用流程································································································2
1.2.1 目標分析·······································································································2
1.2.2 數據準備·······································································································3
1.2.3 特徵工程·······································································································4
1.2.4 模型訓練與調優······························································································5
1.2.5 性能度量與模型應用························································································6
1.3 Python 機器學習工具庫簡介·················································································6
1.3.1 數據準備相關工具庫························································································6
1.3.2 數據可視化相關工具庫·····················································································7
1.3.3 模型訓練與評估相關工具庫···············································································8
小結····························································································································9
課後習題 ··················································································································.10
第 2 章數據準備 ···········································································································.12
2.1 數據質量校驗····································································································.12
2.1.1 一致性校驗·································································································.12
2.1.2 缺失值校驗·································································································.15
2.1.3 異常值校驗·································································································.17
2.2 數據分佈與趨勢探查·························································································.18
2.2.1 分佈分析····································································································.18
2.2.2 對比分析····································································································.22
2.2.3 描述性統計分析···························································································.25
2.2.4 周期性分析·································································································.28
2.2.5 貢獻度分析·································································································.29
2.2.6 相關性分析·································································································.31
VIII
2.3 數據清洗···········································································································.35
2.3.1 缺失值處理·································································································.35
2.3.2 異常值處理·································································································.38
2.4 數據合並···········································································································.39
2.4.1 數據堆疊····································································································.39
2.4.2 主鍵合並····································································································.43
小結·························································································································.45
課後習題 ··················································································································.45
第 3 章特徵工程 ···········································································································.48
3.1 特徵變換···········································································································.48
3.1.1 標準化·······································································································.48
3.1.2 獨熱編碼····································································································.54
3.1.3 離散化·······································································································.55
3.2 特徵選擇···········································································································.58
3.2.1 子集搜索與評價···························································································.58
3.2.2 過濾式選擇·································································································.59
3.2.3 包裹式選擇·································································································.59
3.2.4 嵌入式選擇與L1 範數正則化···········································································.60
3.2.5 稀疏表示與字典學習·····················································································.61
小結·························································································································.63
課後習題 ··················································································································.63
第 4 章有監督學習 ········································································································.66
4.1 有監督學習簡介································································································.66
4.2 性能度量···········································································································.66
4.2.1 分類任務性能度量························································································.66
4.2.2 回歸任務性能度量························································································.68
4.3 線性模型···········································································································.69
4.3.1 線性模型簡介······························································································.69
4.3.2 線性回歸····································································································.69
4.3.3 邏輯回歸····································································································.72
4.4 k 近鄰分類········································································································.75
4.5 決策樹··············································································································.78
4.5.1 決策樹簡介·································································································.78
4.5.2 ID3 算法·····································································································.79
4.5.3 C4.5 算法····································································································.81
4.5.4 CART 算法··································································································.83
4.6 支持向量機·······································································································.86
4.6.1 支持向量機簡介···························································································.86
4.6.2 線性支持向量機···························································································.87
4.6.3 非線性支持向量機························································································.91
4.7 樸素貝葉斯·······································································································.94
4.8 神經網絡···········································································································.98
4.8.1 神經網絡介紹······························································································.98
4.8.2 BP 神經網絡································································································.99
4.9 集成學習···········································································································104
4.9.1 Bagging ······································································································104
4.9.2 Boosting ·····································································································106
4.9.3 Stacking ······································································································115
小結·························································································································116
課後習題 ··················································································································116
第 5 章無監督學習 ········································································································118
5.1 無監督學習簡介································································································118
5.2 降維··················································································································118
5.2.1 PCA ··········································································································118
5.2.2 核化線性降維······························································································121
5.3 聚類任務···········································································································123
5.3.1 聚類性能度量指標························································································124
5.3.2 距離計算····································································································125
5.3.3 原型聚類····································································································126
5.3.4 密度聚類····································································································137
5.3.5 層次聚類····································································································139
小結·························································································································142
課後習題 ··················································································································142
第 6 章智能推薦 ···········································································································144
6.1 智能推薦簡介····································································································144
6.1.1 推薦系統····································································································144
6.1.2 智能推薦的應用···························································································144
6.2 推薦系統性能度量·····························································································146
6.2.1 離線實驗評價指標························································································146
6.2.2 用戶調查評價指標························································································148
6.2.3 在線實驗評價指標························································································149
6.3 基於關聯規則的推薦技術··················································································149
6.3.1 關聯規則和頻繁項集·····················································································150
6.3.2 Apriori 算法·································································································150
6.3.3 FP-Growth 算法····························································································154
6.4 基於協同過濾的推薦技術··················································································159
6.4.1 基於用戶的協同過濾·····················································································159
6.4.2 基於物品的協同過濾·····················································································163
小結·························································································································166
課後習題 ··················································································································167
第 7 章醫療保險的欺詐發現 ··························································································169
7.1 目標分析···········································································································169
7.1.1 背景··········································································································169
7.1.2 數據說明····································································································170
7.1.3 分析目標····································································································171
7.2 數據準備···········································································································172
7.2.1 描述性統計分析···························································································172
7.2.2 數據清洗····································································································172
7.2.3 分析投保人和醫療機構的信息·········································································173
7.3 特徵工程···········································································································177
7.3.1 特徵選擇····································································································177
7.3.2 特徵變換····································································································178
7.4 模型訓練···········································································································182
7.5 性能度量···········································································································184
7.5.1 結果分析····································································································184
7.5.2 聚類性能度量······························································································188
小結·························································································································190
第 8 章中醫證型關聯規則分析 ······················································································191
8.1 目標分析···········································································································191
8.1.1 背景··········································································································191
8.1.2 數據說明····································································································191
8.1.3 分析目標····································································································192
8.2 數據準備···········································································································193
8.2.1 數據獲取····································································································193
8.2.2 數據清洗····································································································195
8.3 特徵工程···········································································································196
8.3.1 特徵選擇····································································································196
8.3.2 特徵變換····································································································197
8.4 模型訓練···········································································································201
8.5 性能度量···········································································································202
8.5.1 結果分析····································································································203
8.5.2 模型應用····································································································204
小結·························································································································204
第 9 章糖尿病遺傳風險預測 ··························································································205
9.1 目標分析···········································································································205
9.1.1 背景··········································································································205
9.1.2 數據說明····································································································206
9.1.3 分析目標····································································································207
9.2 數據準備···········································································································207
9.2.1 數據探索····································································································207
9.2.2 數據清洗····································································································209
9.3 特徵工程···········································································································209
9.4 模型構建···········································································································211
9.4.1 交叉驗證····································································································211
9.4.2 模型訓練····································································································213
9.5 性能度量···········································································································214
9.5.1 結果分析····································································································214
9.5.2 模型評價····································································································216
小結·························································································································216
第 10 章基於深度殘差神經網絡的皮膚癌檢測································································217
10.1 目標分析·········································································································217
10.1.1 背景·········································································································217
10.1.2 圖像數據說明·····························································································218
10.1.3 分析方法與過程··························································································219
10.2 圖像數據預處理······························································································219
10.2.1 圖像預處理································································································219
10.2.2 查看處理後的圖像·······················································································222
10.3 模型構建·········································································································223
10.3.1 捲積神經網絡(CNN) ················································································223
10.3.2 殘差網絡(Residual Network) ·······································································226
10.3.3 ImageDataGenerator 參數說明·········································································228
10.3.4 訓練深度殘差神經網絡模型···········································································229
10.4 性能度量·········································································································231
10.4.1 性能分析···································································································231
10.4.2 結果分析···································································································232
小結·························································································································234
第 11 章基於 TipDM 數據挖掘建模平臺實現醫療保險的欺詐發現··································236
11.1 TipDM 數據挖掘建模平臺················································································236
11.1.1 首頁·········································································································237
11.1.2 數據源······································································································238
11.1.3 工程·········································································································239
11.1.4 系統組件···································································································240
11.1.5 TipDM 數據挖掘建模平臺的本地化部署···························································241
11.2 快速構建醫療保險的欺詐發現工程··································································243
11.2.1 獲取數據···································································································244
11.2.2 數據準備···································································································247
11.2.3 特徵工程···································································································250
11.2.4 模型訓練···································································································253
小結·························································································································255
參考文獻 ·························································································································256